Evaluating health data integrations during a trial

Evaluate a health-data integration by reconciling source records against the account inputs and score shown in the trial platform. Check identity, field meaning, refresh timing and missing values before relying on an alert. A connection indicator does not answer those questions.

Choose one health question and its source

Name the decision the success team expects to make, such as investigating whether a customer's intended workflow has stopped completing. Identify the source record that represents a valid completion, the customer entity it belongs to and the period when the task should occur. Keep customer-reported outcomes and service impact separate from product events; they answer different questions.

Record each input's source system and field, business meaning, customer identifier, source update time, expected arrival time, owner and missing-value rule before importing data. Decide which system may edit it. A precise-looking calculation can still rest on a wrong account mapping.

Reconcile representative cases

Select permitted records that expose likely failures: an ordinary account, a group with two child accounts, a recent source change, a missing input and a record that should be excluded. Compare each source record with its destination.

Trace a mismatch through the mapping and sync rule before changing a score threshold. Keep the observation, correction and person responsible together.

Check the configured connector

Vendor documentation supplies test questions, while the configured trial supplies the result. Gainsight's Salesforce documentation covers account, contact and case data. It describes real-time Salesforce-to-Gainsight sync for those objects but says real-time bidirectional sync is unavailable.

Planhat says changes made in Planhat sync automatically to Salesforce in near real time, while its Salesforce-to-Planhat sync runs every 10 minutes, with an alternative job every 30 minutes. Vitally says Salesforce Accounts can be bi-directionally synced and/or imported into Vitally as Vitally Accounts, with whitelisted fields added to Vitally account profiles. Vitally also says its customer analytics, such as health scores, can be pushed back to Salesforce records using the supported field types it documents.

Make a permitted change at the source and observe its destination. Then check whether a change in the success platform should travel back and under which rule. Inspect the proposed schedule, field permissions, filters and error information. A successful manual refresh does not establish routine behaviour.

Sync Frequency and Bidirectional Capabilities Across Platforms

Inspect the score after a change

Record component values, update times and the missing-data rule before and after changing an input. Vitally documents an overall health score as a weighted average of component scores, and says that when nulls are excluded the health score simply ignores that metric. A changed total can therefore reflect data coverage as well as customer activity.

Ask an account owner to explain the trial view. Can they tell which input changed and when, distinguish a missing feed from a genuine decline and identify who owns the source? Keep any answer that requires administrator help visible in the evaluation.

Health Score Calculation and Null Handling in Vitally

Make an acceptance decision

Classify each required input as accepted, needs repair or unavailable for this evaluation. For an accepted input, retain its mapping, refresh expectation and reconciliation evidence. Repeat the check after a repair. For an unavailable input, decide whether the workflow can operate with a clearly labelled gap or whether the purchase evaluation must wait.